Several low-power experiments have evaluated various methods, including those based on noise analysis, to measure the subcritical reactivity in accelerator-driven systems (ADSs). Similar experiments are planned at the Bhabha Atomic Research Centre (BARC). We have developed a new theory of reactor noise in ADSs taking into account the non-Poisson character of the source. One of the aims of the BARC experiments is to verify the theory and to interpret the results in terms of the new theory. As part of the experimental planning, a simulation of the kinds of results that might be expected with different counting and analyzing setups is necessary. We have developed an analog Monte Carlo code for carrying out these simulations. The simulator generates a detailed time history of counts in the detector so that any method of analysis can be carried out. Since analog Monte Carlo takes a very long computing time, instead of carrying out a simulation to yield results equivalent to transport theory, we attempt to reproduce results equivalent to few-group diffusion theory, which requires much less time. We discuss the basic theory of the simulation method and the results of our simulations on a simplified model of a proposed subcritical assembly.